Most children with Specific Language Impairment (SLI) have mixed receptive and expressive deficits. Even for the children whose impairment appears limited to expression, comprehension status remains in doubt because the tests employed are not sufficiently sensitive and because they tell us little about the processes that ultimately lead to correct or to incorrect endpoint comprehension responses. Previous investigations of SLI have largely relied on spontaneous or elicited production data and on a small number of off-line comprehension tasks. These tasks reveal little about the nature of real time processing and the time course of component processes underlying comprehension. These processes can only be understood by detailing, moment-by-moment, the temporal course of information integration as it takes place during language comprehension. The goal of this project is to examine the nature of on-line lexical processing in children with SLI. A series of on-line experiments are organized into four Groups. Experiment Group A addresses word recognition by examining how (and whether) SLI children resolve segmentation ambiguities. Experiment Group B addresses the organization of the mental lexicon in SLI children by examining lexical access and semantic and phonological priming. Experiment Group C explores how lexical information is integrated into processing at the syntactic level, focusing on properties of verbs (argument structure and thematic roles) and on processing of closed-class function words and their integration into sentence processing. Experiment Group D examines the interaction of lexical processing with the on-line assignment of antecedents to anaphors on the one hand, and with the establishment of relations between moved phrases and their traces on the other hand (as in object relative constructions). In all the experiment series, manipulations of temporal rate, processing load, and prosodic saliency will yield converging evidence regarding the nature of processing deficits. Children with SLI will be compared with two groups of children developing language typically: children matched for chronological age, and children matched for comprehension abilities. The data should yield evidence that will elucidate the nature of SLI and suggest new approaches to assessment and intervention for this disorder.